On Mon, Jul 29, 2013 at 10:09 AM, Albert ARIBAUD albert.u.boot@aribaud.net wrote:
Hi Michael,
On Mon, 29 Jul 2013 08:57:53 -0400, Michael Spang mspang@csclub.uwaterloo.ca wrote:
Albert,
That's not a correct characterization of the bug.
The incoherent cache lines are from before the relocation stage. If U-Boot is relocating from RAM, and later copies the OS there without invalidating those lines, then that's a bug in U-Boot.
Thanks for this pointing out this scenario, which is correct, although it was not raised in the original bug description.
I begs however the question whether anything from re-relocation can survive in the icache from between the moment the relocation starts and the moment the OS is given transfer to. IOW, was this issue actually met?
Yes, I had boot failures without this patch on the device I was porting to (TS-7800). It happened consistently.
Anyway I stand by my statement that even if there is an issue to fix, this patch fixes it in the wrong place.
Ok, that's fair. I was hoping to repost this series someday and will take that into account.
Michael